You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Related to #494 (comment)
Volumes should be deduplicated based on the volume itself, not the name. Then, if we have name conflicts, we should change the volume.name+volumeMount.name (we have to do that when processing a task (ContainerOp), because later we won't know which volume is used by which volumeMount).
One solution to this is to have a method like task.add_mounted_volume(volume, volume_mount) which will automatically set the volume name to volume structure hash. That way different volumes will get different names and same volumes will get same names.
The text was updated successfully, but these errors were encountered:
Related to #494 (comment)
Volumes should be deduplicated based on the volume itself, not the name. Then, if we have name conflicts, we should change the volume.name+volumeMount.name (we have to do that when processing a task (ContainerOp), because later we won't know which volume is used by which volumeMount).
One solution to this is to have a method like task.add_mounted_volume(volume, volume_mount) which will automatically set the volume name to volume structure hash. That way different volumes will get different names and same volumes will get same names.
The text was updated successfully, but these errors were encountered: